The present application claims the benefit of U.S. patent application Ser. No. 16/566,250, filed on 10, 9, 2019, which is incorporated herein by reference.
Disclosure of Invention
It should be appreciated that an electrostatic motor provides a "back current" or back MMF (magnetomotive force) that is substantially similar to the back EMF (electromotive force) of a standard electromagnetic motor. The measurement of MMF, also known as reverse current, is typically a vector with phase and amplitude, one or both of which may provide information for inferring position and or velocity. The present invention provides a method of extracting MMF measurements from an electrostatic motor when the electrostatic motor is advantageously powered by a current drive, such as described in U.S. patent 9,979,323, named by the common inventor Ludois and incorporated herein by reference. Importantly, the present invention allows for easy-to-handle voltage sampling at the motor terminals without the need for bulky current transformers or the like. At medium speeds, the MMF value can be used to infer position and/or motor speed without the need for a resolver with the necessary resolution. At low or zero speeds, when insufficient/insignificant MMF is present, position and/or speed may be sensed by injecting current in the rotor or stator in place of MMF measurements. On the other hand, at high speeds, injecting a current that can be distinguished from the drive current can make MMF sensing more difficult.
In particular, then, in one embodiment, the invention provides an electrostatic motor driver for an electrostatic motor that includes a set of current source drivers adapted to be connected to a plurality of stator electrodes. The reverse current monitoring circuit detects a reverse current value from the electrostatic motor proportional to the rotor speed and the reverse current adjustment circuit receives the detected reverse current value to provide at least one of an estimated rotor position and a rotor speed, which are provided to the comparison circuit which receives the motor control value and at least one of the estimated rotor position and the rotor speed and compares the motor control value to generate an error output to the set of current source drivers. The reverse current monitoring circuit may obtain a voltage measurement at a connection between the current source driver and the corresponding stator electrode.
It is therefore a feature of at least one embodiment of the present invention to provide position and/or speed measurements of a motor shaft of a high pole count electrostatic motor type for low speed and high torque applications using simple voltage monitoring, thereby eliminating the need for direct output reverse current sensing.
The current source driver may provide a set of electrical switches in series with a current source implemented by an inductance for modulating the current to the stator electrodes and regulating the voltage.
It is therefore a feature of at least one embodiment of the present invention to provide a resolver-free position/rotation rate sensor compatible with current source drivers that facilitate this type of electrostatic motor.
The reverse current monitoring circuit may compare the voltage measurement to a common voltage to extract a stator voltage isolated from the common mode voltage.
It is therefore a feature of at least one embodiment of the present invention to eliminate the effect of a highly variable common mode voltage on the calculation of the reverse current.
The monitoring circuit may model the impedance of the stator circuit to infer the current through each stator electrode, and may compare the inferred current to the drive current from the current source driver associated with the stator electrode to infer the reverse current.
It is therefore a feature of at least one embodiment of the present invention to infer forward current into the stator to calculate reverse current from voltage measurements as needed.
The reverse current regulation circuit may also measure a peak value of the reverse current to provide a speed signal, and the comparison circuit may also use the speed signal to provide an error output.
It is therefore a feature of at least one embodiment of the present invention to provide independent measurement of speed useful for speed control of, for example, an electric motor.
The reverse current adjustment circuit may extract an estimated rotor position based on a change in the reverse current.
It is therefore a feature of at least one embodiment of the present invention to infer position from the position-related changes in reverse current.
The electrostatic motor driver may further include a signal generator providing an injection signal to one of the rotor and the stator;
An extraction circuit monitoring at least one of the rotor and the stator to extract a resultant signal indicative of at least one of a capacitive coupling between the rotor and the stator and an effective capacitance that varies according to the salient pole and the at least one of the spatially aligned rotor and stator, and
An adjustment circuit that receives the resulting signal to provide an estimated rotor position;
Wherein the comparison circuit also receives the estimated rotor position signal from the salient pole and spatial alignment adjustment circuit to produce an error output.
It is therefore a feature of at least one embodiment of the present invention to adjust the low signal-to-noise ratio of the reverse current signal at low speed for low speed control.
The electrostatic motor drive may further include a switch for selectively transmitting one of the estimated rotor position signal from the salient pole regulation circuit and the estimated rotor position signal from the reverse current regulation circuit for use by the comparison circuit.
It is therefore a feature of at least one embodiment of the present invention to provide an automatic basis for switching between reverse current and injection current position sensing.
The switch may be controlled by an estimated rotor speed derived from at least one of the reverse current regulation circuit and the salient pole regulation circuit.
It is therefore a feature of at least one embodiment of the present invention to utilize the rotor speed derived from the sensing systems of the present invention to select between these sensing systems.
These particular features and advantages may apply to only some embodiments falling within the claims and thus do not limit the scope of the invention.
Detailed Description
Electrostatic motor design
Referring now to fig. 1, an electrostatic drive system 10 may include an electrostatic motor 12, in one example, the electrostatic motor 12 having one or more disk-shaped plates with radially extending, circumferentially displaced stator electrodes 16. The stator electrodes 16 interact with corresponding radially extending, circumferentially displaced rotor electrodes 20 on a corresponding disc rotor 18 positioned adjacent the disc stator 14. For simplicity, stator poles 16 and rotor poles 20 are shown on the visible surfaces of stator 14 and rotor 18, however, stator poles 16 and rotor poles 20 are typically in close proximity on opposite sides of stator 14 and rotor 18. This type of motor will be referred to as an "axial field" motor, generally referring to the alignment of the electrostatic field along the axis of rotation 25 of the rotor 18.
The present invention also contemplates operation with a "radial field" motor having an electric field extending perpendicular to axis 25, such as a circumferentially nested cylindrical plate or axially extending key ring. Typically, the number of rotor electrodes 20 in each of these types of motors will match the number of poles of the motor. The number of poles will typically be in excess of 16, more typically in excess of 60, preferably 96 or more.
Axial field motors and radial field motors are described in U.S. patent 9,184,676, 2016/0211775, and 2016/0344306, all assigned to the assignee of the present invention and incorporated herein by reference. The invention is applicable to both types of motors.
In both radial flux and axial flux motor designs, rotor 18 may be supported for rotation on a drive shaft 24 extending along an axis 25 to extract mechanical work. A slip ring or brushless power transmission system 22 (e.g., capacitive or inductive) is attached to the drive shaft 24 such that power from a stationary rotor power supply 26 can be conducted to the rotating rotor electrode 20, as is commonly understood in the art, to provide electrostatic polarization of the rotor 18.
Overview of variable speed drives
The variable speed drive 32 may provide a controlled application of power to the stator electrodes 16 of the stator 14 based on the position signal generated by the position detection system 30. In this regard, the variable speed drive 32 may receive a command signal 34, such as a position, speed, torque, or other related quantity, and determine an appropriate variable current to apply to the stator electrode 16 to provide operation of the electrostatic motor 12 in accordance with the command signal 34. Thus, the output of the variable speed drive 32 will provide a plurality of phases 36 (also designated A, B, C for three-phase embodiments) associated with different stator poles 16, thereby providing those stator poles 16 with sinusoidal or other continuously varying signals required to control motor operation.
Referring now to fig. 2, the instantaneous value of the output phase 36 required for a given command signal 34 may be a function of not only the command signal 34 but also the rotor position and characteristics of the motor 12. This processing required to generate output phases 36 may be simplified by coordinate transformations known in conventional electromagnetic motors, wherein a plurality of phases 36 that are constantly changing are mapped to a frame of reference that rotates with motor rotor 18. This reference frame is referred to as the d-q reference frame in which the d-axis (the direct axis) is aligned with the positive electrode on electrode 20 and the q-axis (the orthogonal axis) is positioned at 90 degrees to the d-axis. Looking in this frame of reference, the complexity of the waveforms at the multiple phases 36 (e.g., in a three-phase system, referred to as A, B, C) translates into a single vector that is largely unchanged for steady-state operation of the motor 12. Details of this transformation in the context of electromagnetic machines are described, for example, in "Vector Control AND DYNAMICS of AC Drives" oxford university press, 1996, 1 st edition (including pages 88 to 102), of d.w. novotny and t.a. lipo, where basic mathematics is also applicable to the present invention.
Using this transformation, the present invention provides feedback control of the current source driver 40 such that the phase 36 is connected to each stator electrode 16. In this aspect, the voltage from each of the phases 36 is measured and the measurements are received by the ABC-dq conversion circuit 42. The ABC-dq transformation circuit 42 also receives a position signal 44 and a velocity signal 45 from the position detection system 30 to convert the received phase signals (A, B, C) into vectors in d-q space, referred to as "measured" d-q vectors 48.
The input command signal 34 will be converted to a similar "desired" d-q vector 50 by an input conversion circuit 52. When the electrostatic motor 12 is not operating in steady state, the desired d-q vector 50 will typically have a different angle and a different magnitude than the measured d-q vector 48. When the input command signal 34 is a torque value, the magnitude of the desired d-q vector 50 will be proportional to the desired torque, and the desired angle relative to the q-axis will depend on the type of motor 12. For a non-salient machine, the angle would simply be zero degrees (the desired d-q vector 50 is aligned with the q-axis), however, for a salient machine, the calculation would be more complex, as discussed in U.S. patent 9,979,323 assigned to the assignee of the present invention and incorporated herein by reference. The ideal angle is the angle that provides the maximum torque per voltage to reduce motor losses. Alternatively, the command signal 34 may be a speed value, in which case the speed signal 45 is used. More generally, any control strategy may use both the position signal 44 and the velocity signal 45.
Once the desired d-q vector 50 is determined, the desired d-q vector 50 is compared to the measured d-q vector 48 to produce an error value 53 at a comparison circuit 54 that controls the current source 40. In the simplest case, the error value 53 is simply the difference between the desired d-q vector 50 and the measured d-q vector 48, however, alternatively the difference may be further processed, for example under a proportional/integral/derivative type control strategy, wherein the error value 53 is a weighted combination of the differences, a time-running integral of the differences and a derivative of the differences. It should also be appreciated that other control strategies may be used by the comparison circuit 54, including feedback and/or feedforward of other measured variables derived from the motor 12.
Still referring to FIG. 2, the error value 53 is then provided to the dq-ABC transform circuit 56 (as an inverse transform) operating in the opposite direction to the ABC-dq transform circuit 42 to change the error value 53 as a vector in d-q space to the phase 36 in the non-rotating system.
This feedback control process of traversing the loops of the ABC-dq conversion circuit 42 and the dq-ABC conversion circuit 56 continues during operation of the motor 12.
When the command signal 34 is a different value, such as a desired rotational speed (e.g., RPM), an additional optional feedback loop may be incorporated, such as at optional comparison block 58, using the position signal 44 for inferring speed and using the difference between the desired RPM and the inferred RPM of the command signal 34, to create a torque value, which may then be processed as discussed above with respect to the torque signal. Other input signals may also be processed in this manner, and in this regard, the present invention contemplates that programmable command signals 34 may be used for different states, such as soft start and soft stop of motor 12, and different motor RPM or operating conditions.
The ABC-dq conversion circuit 42, the input conversion circuit 52, the comparison circuit 54, and the dq-ABC conversion circuit 56 may be implemented by discrete circuitry or, preferably, by a high-speed computer processor executing programs stored as firmware in non-transitory computer memory, for example, and operating in the digital domain with analog-to-digital converters.
Referring now to fig. 3, the practical implication of complex field control of an electrostatic motor is achieved by the ability to generate a "stiff" current output signal at the power level required to drive the electrostatic motor 12, i.e., an output that can provide open loop current control in the face of rapidly fluctuating voltages at the multiple phases 36 caused by changes in capacitive coupling as the motor 12 rotates. The present invention contemplates that the electrostatic motor 12 will operate at a power in excess of 10 watts, typically in excess of 100 watts, desirably in excess of 1000 watts.
The necessary "current source" output can be generated by using one or more series inductive elements 78, taking advantage of the inductive characteristics of the current flowing through the inductor, the cumulative characteristics of self-inductive energy within the inductor magnetic field, and the resistance to changes in current. The present invention recognizes that this characteristic can be utilized to provide sufficient output current stiffness to enable the output voltage to be regulated without impeding dynamic control of the current necessary for "charge steering" control or variable speed capability of the motor. In this regard, the magnitude of the inductance must provide current regulation (and thus energy storage) at the desired motor power level to provide, for example, control of the current output to the motor to within 25%, typically within 10%, desirably within 5% of the commanded value controlling the semiconductor switches. The construction of such a current source driver is described in U.S. patent 9,960,719 assigned to the assignee of the present invention and incorporated herein by reference.
In one implementation, a DC power source is provided to a set of solid state switches 72, e.g., transistors such as MOSFET transistors, to receive the ABC current values from the switch logic 73. The solid state switches 72 are configured, for example, in an H-bridge in which each of the phases 36 is connected to a junction between a pair of series connected switches 72, which in turn bridge a positive power rail 74 and a negative power rail 76, providing a direct current that is stabilized by the inductor 70. The primary use of this circuit may produce a square wave output, however, the present invention contemplates that the phase 36 produced is a continuous waveform of any shape and frequency determined by the control algorithm. Thus, the switch 72 will receive a control signal that determines its switch state, which is pulse width modulated (or modulated by a similar modulation technique including pulse density modulation, etc.). In pulse width modulation, the on-time of switch 72 is varied to determine the average current value output through phase 36. In such modulation, for energy efficiency, the switch 72 is operated in a switch mode (on or off), but is switched at a high rate to produce a continuous waveform (e.g., sine waves of different frequencies) that is smoothed by the capacitance of the electrostatic motor 12. In pulse width modulation, the switching speed of the semiconductor is many times the fundamental frequency of the waveform of phase 36, and typically exceeds 10 to 20 times that frequency.
An inductor 70 may be placed in series with the switch 72 of the H-bridge to stabilize the DC bus feeding the switch 72. Other placement of inductors (e.g., one on each of the phases 36) or use of transformers with leakage inductance may provide similar effects.
Position sensing
Referring again to fig. 2, the position signal 44 and the speed signal 45 may be obtained from a resolver, however, in the present invention, these signals may be provided by the position detection system 30 receiving the voltage signal 90 from each of the phases 36 of the current source driver 40. The position detection system 30 may include two distinct components, a reverse current or "reverse MMF" (magnetomotive force) detector system 93, which includes an MMF detector 92 and a conditioning circuit 120, and an injection current system 131, which includes a current injection circuit 130 and a conditioning circuit 144. Both systems receive voltage measurements of the phase 36 provided to the electrodes 16 of the stator 14 to generate a position signal and a velocity signal.
The reverse MMF detector 93 detects a reverse MMF that is a function of rotor speed and may also be used to provide a position signal based on the MMF's variation with rotation.
Referring now to fig. 8 and 9, mmf detector 92 may measure the voltage at each phase 36 relative to a common voltage reference 94 (e.g., ground) to provide an original phase voltage 96 associated with each phase (e.g., in example 3 phase motor, VAG is the voltage between phase a and ground, VBG is the voltage between phase B and ground, and VCG is the voltage between phase C and ground). These raw phase voltages 96 will include a common mode voltage that is highly variable and can obscure the desired reverse current measurement.
Thus, and referring to fig. 10, each of the raw phase voltages 96 may be combined to extract the isolated phase voltage of each electrode 16. The extraction process can be graphically understood by envisioning the isolated phase voltages as phasors 98 extending at equal angles from a common voltage center 100 and rotating around the common voltage center 100, the common voltage center 100 varying with common mode voltage. It should be appreciated that under the constraint that the phasors 98 must be at equal angles to each other, the length of each phasor 98 (isolated phase voltage) can be uniquely calculated from knowledge of the length and relative angle of the phasors 96 using geometric analysis, thereby eliminating the effects of common mode voltages.
Referring now to fig. 8, each of the stator electrodes 16 may be modeled by a fixed capacitance 102, a fixed resistance 104, and a current source 106 representing the inverse MMF as a function of rotor speed. The capacitance 102 will typically vary as a function of rotor position, but may be modeled as an empirically determined average value, and the position variation is due to the current source 106. The capacitance 102 and the fixed resistance 104 may be empirically determined or may be inferred during operation of the electrostatic drive system 10.
The model may be used to determine the inverse MMF of the current source 106 by applying an isolated phase voltage (e.g., VA) to the model to determine the receive current 108 (combined current through the capacitor 102 and resistor 104) that would occur if a measured voltage (e.g., VA) were applied across the model. The received current 108 may then be compared to a command current 110 from the current source driver 40. The difference between current 110 and current 108 will be the effective current from current source 106 as the reverse MMF.
Referring now to FIG. 6, the calculated reverse MMF current will vary over time as the capacitance 102 actually varies as the rotor 18 rotates to generate the reverse MMF signal 115. However, the amplitude 112 of the calculated inverse MMF signal 115 will be proportional to the rotational speed of the rotor 18 and thus can be used to determine the rotor speed. The variation of MMF signal 115 over one cycle 114 provides an indication of the position of rotor 18 and the rotational distance of each cycle 114 will be equal to 360 ° of rotational travel divided by the number of poles (three in this simplified case) of motor 12. It will be apparent that angular positions less than one period 114 may also be determined from regular voltage variations during the period 114.
In general, the position signal from each phase may be transformed into a d-component and a q-component, where in fig. 6 the d-component is shown in solid lines and the q-component is shown in dashed lines and is simply represented as the length of the corresponding quadrature-phasors.
Referring now to fig. 4, the inverted MMF signal 115 may generally be processed by a conditioning circuit 120, such as providing bandpass filtering 122 to extract a noise-reduced MMF signal 115 that may be provided to a mapper 124, the mapper 124 mapping, for example, the voltage value within one period 114 to a particular angle value as the position signal 44, and a peak follower 126 extracting the amplitude 112 for use as the velocity signal 45. Other well-known signal conditioning techniques may be used, including, for example, constructing a viewer that fits the data to a model or the like.
Referring to fig. 2 and 5, an alternative source of position information may be obtained by current injection provided by current injection circuit 130. Typically, the current injection circuit 130 may create a high frequency injection signal by the injection signal generator 132, e.g., the high frequency injection signal has a frequency at least 10 times the period 114. The injection signal generator 132 may provide an injection output 134, and the injection output 134 may be added to the output of the transformation circuit 56 to superimpose an additional current signal on one electrode 16 of the stator 14 through the current source driver 40.
The injection signal may be used in two ways. The first method uses the injection signal to measure the capacitive coupling between the stator 16 and the rotor 20, e.g., as the rotor 20 rotates. In this case, a voltage signal 136 induced through injection output 134 on the stator may be received by rotor 118 but modified by changing the mutual capacitance between rotor 18 and stator 14 as rotor 18 rotates. The signal 136 may be received by a high pass filter 139 for reducing noise content and then demodulated using an extraction circuit such as a demodulator 138 (depicted schematically as a rectifier 141 and a low pass filter 143) to extract the envelope of the signal 136 having the modulated frequency. The modulation frequency will have a period representing a frequency corresponding to the rotational speed of the rotor 20 and may thus be used to determine the rotor speed 45, e.g. using the frequency detector 145, e.g. measuring the period and reversing it. The envelope of the phases may be used to provide a position signal 44 measurement in a manner similar to that described above with respect to fig. 6. These output signals 44 and 45 may again be processed by a conditioning circuit 144, for example to provide filtering or more complex signal conditioning using observer techniques or the like.
Referring briefly to fig. 11, it should be appreciated that the injection process may be reversed, wherein the injection circuit 130 injects directly into the rotor 20 and then monitors the resulting change in the signal 90. In this case, the signal 134 shown in fig. 2 is not required.
As an alternative to measuring the above-described capacitive coupling variations, the injection signal may be used to detect salient pole variations of the electrostatic motor 12. Referring also to fig. 7, in this case, the voltage signal 90 may be monitored by the salient pole detection circuit 131 to detect a load change of the injection signal from the current source driver 40 caused by a salient pole change of the stator 16. The voltage signal 90 may be received by a saliency circuit 131, which saliency circuit 131 provides the same functional components as the current injection circuit 130, including a high pass filter for reducing noise content and a demodulator and low pass filter 143 for extracting the envelope 150 of the signal 136 having a modulated frequency. The modulation frequency will have a period 152 representing a frequency that is typically twice as fast as period 114 and thus may be used to determine the rotor speed 45, e.g., using a frequency detector 145, e.g., measuring period 152 and reversing it. The envelope 150 of the phases may be used to provide position signal 44 measurements in a manner similar to that described above with respect to fig. 6. The output signals from the salient pole circuit 131 or the current injection circuit 130 may be selected by the switch 161 to be used as the output signals 44 and 45.
Referring again to fig. 2, each of the MMF detector 92 and the current injection circuit 130 or the saliency circuit 131 may provide both a position signal and a velocity signal, however, the MMF detector 92 has a poor signal-to-noise ratio at low rotor speeds and, thus, at low rotor speeds, the current injection circuit 130 or the saliency detection circuit 131 may be used to provide both a position measurement and a velocity measurement. On the other hand, when the motor 12 is moving at a high speed, excellent measurement results provided by the MMF detector 92 can be used.
In this regard, the switching circuit 160 may automatically select between the output from the regulating circuit 120 and the output from the regulator 140 based on the speed signal obtained from the comparing circuit 54. In this regard, the comparison circuit 54 switches between the position detection systems of these different detection systems depending on the speed of the rotor 18.
It should be appreciated that the present invention provides the ability to properly control the voltage vector applied to the electrostatic motor by closed loop voltage regulation, thereby also providing the ability to control torque, and in this way provide torque control.
Certain terminology is used herein for reference purposes only and is therefore not intended to be limiting. For example, terms such as "upper," "lower," "above," and "below" refer to directions in the drawings to which reference is made. Terms such as "front", "rear", "bottom" and "side" describe the orientation of portions of the component within a consistent but arbitrary frame of reference which is made clear by reference to the text and the associated drawings describing the component under discussion. Such terminology may include the words above specifically mentioned, derivatives thereof and words of similar import. Similarly, the terms "first," "second," and other such numerical terms referring to structures do not imply a sequence or order unless clearly indicated by the context. Although the stator and rotor are shown as disks in the disclosed embodiments, it is not required that the stator or rotor be in the form of disks.
When introducing elements or features of the present disclosure and the exemplary embodiments, the articles "a," "an," "the," and "said" are intended to mean that there are one or more of the elements or features. The terms "comprising," "including," and "having" are intended to be inclusive and mean that there may be additional elements or features other than the specifically indicated elements or features. It is further to be understood that the method steps, processes, and operations described herein are not to be construed as necessarily requiring their performance in the particular order discussed or illustrated, unless specifically identified as an order of performance. It should also be appreciated that additional or alternative steps may be employed.
In particular, the invention is not intended to be limited to the embodiments and descriptions contained herein, and the claims should be construed as including modifications to these embodiments that include both part of the embodiments and combinations of elements of the different embodiments within the scope of the appended claims. All publications, including patent publications and non-patent publications, described herein are incorporated by reference in their entirety.